Plus, I just don't understand all the NIH. Why isn't this kind of stuff being developed in one place and then forked by the respective countries instead of everyone creating their own tiny island?
That said, I don't have any major gripes with how the code turned out, just with the cost.
I have followed this project closely. Checked the protocols. The APIs. And what‘s built on top.
The only complex thing I can think of is the way cases are confirmed. That‘s a bit involved. But it was worked out in advance.
When I checked the actual Swift code was just a bit less than 10k.
The backend apps are fairly small Spring boot apps.
There was a messy attempt before (not finished in time at all) and they fixed and finished that attempt.
This boggles my mind. Sure, it will take a reasonable effort to get widespread use of these apps. But the impact would be huge. Even low rates of app use have a real chance of lowering the reproduction rate significantly, and reasonably broad usage bring it less than 1.0.
The technology is designed in the right way, where anonymous data is being collected with no app required. People can download the app right after they get a positive test and report themselves positive retroactively. Same on the other side, you can download the app and know how much contact you've had with positive cases retroactively.
I'm no longer surprised the US federal government is doing nothing to push this. But I expected more out of the states.
[0] https://9to5mac.com/2020/07/13/covid-19-exposure-notificatio...
It boggles my mind that I have to remind many app publishers that we live in a globalized world and that it’s a natter of a few minutes work to just select the countries to make the app available in. More so for free apps where the publisher doesn’t have to worry about taxes and payments.
Even for apps that are not localized to English or other languages, the publishers should prefer to get them out to as many people as possible in the whole wide world who may be visiting or on a long stay in a specific country.
In this case, the app developer can surely include help text that explains where the app is meant to be used or would be useful (it’s also not like people with a German App Store account living in some other country would be able to use this app effectively).
One could build 100 apps with server code and docs with that money, and still have a few million left over for hookers and blow.
Stop defending corruption.
100 random apps or app that have to handle pandemic in country?
Also what about support and infrastructure to handle those milions of concurrent users
Same for what you call concurrent users. Those users have no identity at all and essentially need to anonymously download a bunch of static files which are the same for everyone from a bucket periodically. Put a CDN in front. Done.
The contract tracing apps are really basic... a step above torch apps.
(And the MSS/GRU/NSA/Mossad will all insert their code into your app bundles without you needing to do anything...)
The only backend touches that apps do is a few login/register APIs when you first launch it, and if you ever test positive and agree to send your data it sends a big json blob to an AWS Lambda nodejs function that pretty much just uploads the json to S3.
I don't recall how complex the login/register stuff is, but I don't remember thinking "That's gonna need heroics to register 10 or 20 million users quickly" when reviewing the code.
(If you wanna go look yourself, try here: https://github.com/opentrace-community )
The loss in public reputation if they had failed was not worth even participating.
Also, having worked in a corporate setting would have told you that you can't launch anything for less than 1m EUR/USD, can't launch two things for less than 5m and 3 for less than 20m.
20m really is a laughable sum.
The government was then talking it up with "If enough people download the app, we'll let you go to football games again!!!"
Current media reports say none of our local health departments have found _any_ contacts using it that they hadn't already found via manual contact tracing.
https://www.abc.net.au/news/2020-07-15/coronavirus-app-limit...
(I think 100 apps+backends for $10mil is a bit optimistic, if those backends are required to be HA and support a significant percentage of a countries population. As a single datapoint, where I work has many iOS+Android+backend app projects for gov departments that serve tens to hundreds of thousands of users that cost them high 5 or low 6 figures plus 4 figures a month support/hosting, but if they wanted me to support the 60% of the population (~15mil users) that our CovidSafe app was targeting it'd most likely come to more like mid to high 6 figures, and certainly 5 figures a month to host/support it. So I'd have said maybe 20 apps realistically for that price. Still just as fraudulent a rorting of taxpayer money though...)